Generalisation from cases to theory 63 

The three cases have specific similarities and differences that provide a useful base for testing the theoretical propositions in this research. The three firms are all already exposed to international R&D environment for periods longer than a decade. They have all established a degree of technological leadership in their industries at a global scale. They have all established strong R&D capability at home. Findings relating to theoretical propositions are outlines next.

Hypothesis 1: South African firms move core parts or their entire R&D to technologically advanced countries as a result of greater exposure to R&D internationalisation.

Evidence in all three cases refutes Hypothesis 1. The three firms have not relocated their core or their entire R&D to technologically advanced countries abroad as a result of increased international exposure. They have, however, extended certain functions of their R&D to abroad in order to undertake entirely new activities. In doing so, they were motivated by the need to access new knowledge and human capital and exploit their existing capabilities in new markets.

The three firms have already established strong R&D capabilities at home and have extensive experience operating in the international environment. Events in the past two decades, namely the economic globalisation process and South Africa’s economy opening up, have facilitated their increased exposure to global economic factors. Empirical evidence to support this are threefold, namely the deepening role of South Africa in FDI inflows and outflows, international trade in high technology and the international patenting activity.

With increased exposure, R&D activities of local firms have become sensitive to factors in the international economic environment. In “A’s” case, poor demand conditions in the downstream industry globally depressed revenues and profitability, thus reducing internal funding resources for R&D. Equally, these pressures also increase demand for mining technology as companies must innovate and improve productivity. In “B’s” case, new opportunities for R&D exploitation have opened in developing countries, including South Africa, as user industries seek to modernise and improve efficiency. For this firm, there is a strong case for sourcing technologies from abroad with an idea to adapt them to meet local and regional technology gaps. Further pressure arises due to short product life cycle in some of its industries.

Having strong R&D, design and development capabilities at home confirm evidence that the three firms are embedded in the South African innovation system. “B’s” vehicle tracking and digital broadcasting technologies were developed for South African conditions with relatively poor infrastructure. These two technologies are easy to adapt in countries with better infrastructure at reduced specifications. This is advantageous for “B” in international markets. “A’s” proximity to raw material sources and primary

users of its technology (i.e. diamond mining operations) in the Southern Africa region make a strong case for retaining the core R&D at home. The R&D that the three firms perform abroad is for complementing domestic activities rather than replacing it. These firms can increase their R&D investment locally given more funds and readily available capabilities locally. R&D is considered crucial to maintain competitiveness both locally and internationally.

Hypothesis 2: South African firms change their orientation for exploiting

R&D as a result of increased exposure to R&D internationalisation.

Evidence confirms Hypothesis 2. Increased internationalisation has significantly altered R&D exploitation in all cases.

Three points explain this finding. Firstly, in all cases, there is sharper focus and even a higher preference for funding/performing applied R&D and technology development than exploratory fundamental research. In “A’s”, case, the funding pressures reoriented a focus towards “doing less, better with less”. In “B’s” case, priority is afforded to technologies that demonstrate the potential to generate funds for R&D reinvestment and for increasing IP revenues. The discoveries by “C” initiated a range of R&D projects based on the discovery of gas-to-liquid technology, which generated a global appeal. The technologies arising from this discovery could only be applied where the gas resources are located, in the Middle East.

Secondly, there is increased focus on incremental improvements to existing technologies, discovering new applications of existing technological platforms and creating new markets for them. Increased competitive pressures and reduced funding forced “A” and “B” to enhance R&D efficiency, in ensuring that the R&D funded has better potential for success. However, each of them follows different paths and generates different experiences. “C’s” distinctions between cost-driven and innovation-driven technologies/products prioritisation of effort for different markets, e.g. adaptation research for speciality chemicals (innovation-driven products) in Europe is well-placed because of diverse and demanding customers in automotive design and manufacturing. This confirms that EMNEs, to some extent, differentiate

approaches for R&D exploitation on the basis of location-specific advantages of host locations abroad. There is evidence of fine slicing of activities in the R&D portfolio in “B” and “C” but not as per theoretical proposition in hypothesis 2. Core traditional R&D in both cases is retained at home, while they initiate entirely new activities abroad in locations that offer specific advantages. In these two cases, there is also evidence of scaling up of knowledge exploitation, both at home and abroad, arising from learning internationally, tapping on knowledge signposts they established abroad.

Thirdly, in cases “B” and “C”, there is stronger evidence for scaling up of R&D exploitation through internationalisation. This is achieved by tapping onto knowledge ‘sign-posts’ these firms have established abroad. “B” benefits from backward linkages with OEMs abroad and from forward linkages to customers/markets locally and in developing markets it serves. In “C’s” case, a speciality chemical example cited above is relevant here.

Fourthly, in two cases, “B” and “C”, deliberate strategies for international IP exploitation are in place. The two incur costs and effort for patenting abroad is carefully selected jurisdictions in order to maximise IP value, and protect it. With “B” specifically, new structures have been established to significantly alter its historical approaches to R&D exploitation. “B” has established an IP Management and Commercialisation Office in 2011 to actively manage IP as a driver of company’s growth internationally and the Project Management Office (PMO) to monitor all R&D projects and their outputs with a view to monetise and maximise value for the company. A network of Chief Technology Officers (CTOs) and technical people located in various value-based streams supports these two functions. The two functions are centralised in Johannesburg but operate with a global focus.

Rival explanations are noted with respect to R&D exploitation. The three cases demonstrated that EMNEs have different experiences in R&D exploitation abroad. The approaches are influenced by company-specific factors, existing and potential sources of knowledge and the markets served locally and abroad.

Applying the framework proposed in Von Zedtwitz (2005) – Figure 1 (page 8), the three points are deduced.

Firstly, we focus on Type 3. While “A” and “C” are EMNEs exploiting R&D in advanced economies, the challenges anticipated of EMNEs (of being of small size versus competitors, lacking resources and management experience in advanced country markets, and being an unknown brand) do not apply because the two firms are already global leaders in their industries. These two firms’ conditions are more applicable to Type 1 instead even though they originate from an emerging economy. This research deduces, therefore, that the challenges listed above may apply to firms that are new in the international R&D scene that lack specific innovation and output capabilities (Awate et al., 2012).

Secondly, “B’s” expansionary approach on volume-based technology stream is applicable to Type 4. This specifically related to its R&D exploitation of technologies that have been successful locally being deployed/adapted in developing countries.

Thirdly, there are instances where, in each case, they establish collaborations/partnerships abroad to bring technology into South Africa. This fits into quadrant Type 2.

Case specific contexts must be taken into account, however. For instance, “A” and “C’s” operations mainly service their groups globally while “B” competes for customers in an open environment. Extending this specific study to cover cases “A” and “C’s” at parent group level could illuminate the role of M&A, new establishment and reorganisation at group level and motives as well as overall impact on R&D investment and exploitation at that level. Furthermore, a similar research based on a sample of firms in the same industry can reveal industry level implications of R&D internationalisation.

6. Chapter 6: Summary and conclusion